On 12/12/14 7:54 AM, Chris Hegarty wrote:
On 12/12/14 15:40, Pavel Rappo wrote:
Could you please review my change for JDK-8066642?
http://cr.openjdk.java.net/~prappo/8066642/webrev.00/
The first change, to remove SocketSecurityException seems benign.
SocketSecurityException is a subclass of ExportException, so
CommunicationException will continue to be thrown.
The second change, to replace RMISecurityManager with SecurityManager, will
provide equivalent behavior, RMISecurityManager extends SecurityManager but does
not add any state or behavior, but the runtime type of the security will be
changed.
I've taken a quick look through the JDK source and cannot find any places where
the type of the security manager is checked to see if it is an instance of
RMISecurityManager. I also quickly looked in GrepCode, and could not find any
either. Given this, I think the change should be ok.
Regarding RMISecurityManager, yes, there should be no problem replacing it with
an ordinary SecurityManager. There should be no functional difference introduced
by doing so. If anyone runs across code or behavior that differs between
SecurityManager and RMISecurityManager, please let me know; it is most likely an
error!
s'marks
